NEFROLOGIA. Vol. XIV. Núm. 4. 1994 FORMACION CONTINUADA EN NEFROLOGIA Apoptotic cell death in renal disease A. Ortiz-Arduan and E. G. Neilson Penn Center for Molecular Studies of Kidney Diseases, Renal-Electrolyte and Hypertension Division of the Department of Medicine, the Graduate Groups in Immunology and Cell Biology at the University of Pennsylvania, Philadelphia, PA 19104 Understanding the mechanisms of physiologic cell death (apoptotic or programmed cell death) may provide new diagnostic opportunities and, perhaps, additional therapeutic approaches to the clinical management of renal disease. The availability of advanced techniques in cellular and molecular biology to study the genes that regulate apoptosis has resulted in an exponential growth of information, especially in the fields of immunology, oncology, neurology and development 1-4.. We will review the concept of apoptosis, the genes involved in its regulation, its role in physiology and its possible participation in renal disease based on the international Iiterature and our own personal experience. Apoptotic and programmed cell death Only a limited number of things happen to cells resident in organ tissues. They can remain stationary and support the structure-function relationships of the organ, they can protiferate to reproduce themselves, they can sometimes hypertrophy, or they can die. Complex organisms need a physiologic mode for cell death, so as to keep constant their number of cells and eliminate those damaged or that are no longer necessary. Although the concept of physiologic cell death is quite old, renewed interest in the subject dates to the creation of the term apoptotic and the description of its morphological characteristics by Kerr and colleagues in 1972 5. The functional concept of programmed cel l death implies an active participation of the cell in its own n Correspondencia: Eric C. Neilson, MD. C. Mahlon Kline Professor of Medicine. 700 Clinical Research Building. University of Pennsylvania. 422 Curie Boulevard. Philadelphia, PA 19104-6144, USA. death (cell suicide) through the activation of a genetic program. In general programmed cell death has the morphologic characteristics of apoptotic, although there are exceptions. In fact, there is functional, morphologic, and genetic evidente of heterogeneity in this process 6-11, and there are unanswered questions about the physiologic relevance of this diversity. Moreover, there are modes of cell death which may share characteristics of apoptotic and necrosis 12. With this caveat in mind, and given the scarce knowledge of nonapoptotic modalities of programmed cell death, we use the term apoptosis when the characteristic morphology and pattern of DNA degradation have been demonstrated. However, the pragmatist may define programmed cell death as a process that can be modulated through interference with cell death related genes, independently of the morphology or pattern of DNA degradation. A p o p t o t i c is an active process which may be prevented or delayed by the administration of growth factors, certain drugs or by gene manipulation. Cells committed to an apoptotic pathway may be rescued by therapeutic maneuvers 13. Thus, apoptotic offers the opportunity for a therapeutic intervention and this explains the attempts to discriminate between necrosis and apoptosis as a mode of cell death in renal disease. Necrosis, as a mode of cell death, is a passive process in which the cell swells, loses the integrity of the cell membrane, releases toxic and proinflammatoy products, and the DNA is degraded in random manner. Both apoptosis and necrosis, however, can occur at the same time in the same tissue 14. The occurrence of either may depend on the intensity of the precipitating events. For example, tissue ischemia may kill cells by either necrosis or apoptosis 14. The proportion of cells dying by each mechanisms may vary from individual to individual, and the segment of the cell population predetermined to die by apoptosis might be rescued by interference with the genetic program of apoptotic. 391 A. ORTIZ-ARDUAN Y E. G. NEILSON. Characteristics of apoptotic Apoptosis and necrosis may be differentiated by morphologic and functional criteria (table I) 2. From a morphologic view, apoptotic cells display decreased cell and nuclear size with chromatin condensation, preservation of organelles, detachment from adjacent cells, membrane cell blebbing, and fragmentation with the formation of membrane bound bodies, the so-called apoptotic bodies. Some of these features may be appreciated by light microscopy, others are more evident with electronic microscopy, or the use of nucleic acid dyes, Iike propidium iodide and acridine orange. There are several protocols for the identification of apoptotic cells based on changes in cell size or the pattern of nucleic acid staining by cytoflurography 15. Figure 1 .-Internucleosomal DNA fragmentation in renal cells. Tubular epithelial cells were cultured for 24 h in serum free medium in the absence (left lane) or presence (right lane) of 30 ng/mL recombinant murine TNFa. Detached cells were lysed, DNA was isolated, separated in a 7.5% agarose gel and stained with ethidium bromide. Note the characteristic DNA ladder, suggestive of apoptotic, in the TNFcx-treated samples. Table I . Characteristics of apoptotic l l l l l l Physiologically active. Preventable and possibly reversible*. Typical morphology. Integrity of cell membrane. Phagocytosis by adjacent celIs. Internucleosomal DNA degradation. * At least as a process in a cell population. Functionally, apoptotic differs from necrosis in several ways. In apoptosis, the integrity of cell membrane is preserved, and there is no leakage of pro-inflammatory molecules. There is a characteristic pattern of internucleosomal DNA degradation, caused by the activation of a calcium- and magnesiumdependent endonuclease, which results in the formation of multiple DNA fragments with lengths that are integer multiples of 180-200 bp. On gel electrophoresis these fragments are seen as a DNA ladder (figure 1). Other patterns of DNA degradation may occasionally be observed. Apoptotic cells also express new cell membrané structures, Iike immature glucidic structures, phosphatidyl serines and glycoproteins, which determine the rate of recognition and phagocytosis by adjacent cells, all in the absence of severe inflammation. A tissue transglutaminase is activated, which crosslinks cytoplasmic proteins. A p o p t o s i s is an active process, which requires energy in the form of ATP, and the expression or suppression of certain genes. The intracellular transduction pathways of apoptosis are poorly understood, and may involve calcium and inositol-phosphates or changes in protein phosphorylation. Oxidative stress also appears to have a causative role 16. The inhibition of mRNA o r protein synthesis m a y cause or prevent apoptosis, depending on the bala nce between 392 lethal and protective factors in the cell. From this point of view, three modes of apoptosis can be distinguished 2: l By induction: the stimulus which induces apoptosis activates lethal genes, and cycloheximide, an inhibitor of protein synthesis, prevents apoptosis. l By release: the genes promoting apoptosis are already activated, and short lived proteins antagonize them. Macromolecular synthesis inhibitors decrease the Ievels of protective factors and induce apoptosis. l By transduction: the synthesis of new proteins is not necessary, and the occurrence of apoptosis is not affected by cycloheximide. Apoptosis usually affects individual cells, its tissue distribution is patchy and asynchronous, and once initiated, progresses rapidly. It has been calculated that the half-life of the apoptotic cell is a few hours, and it may be phagocytosed in less than one hour. A low percentage of apoptotic cells visible in a tissue section (e.g. less than 0.8%), therefore, may be misleading. It has been suggested that such a low percentage may, nevertheless, result in the loss of up to 50% of the non-required cell mass during the development of the central nervous system 17,18. It has been demonstrated that apoptosis may also attenuate unneeded metanephric mesenchyme following induction'by the ureteric bud 19. The rapid disappearance of the cellular debris and the lack of serious inflammation may make even large scale apoptosis histologically inconspicuous. The lack of easily detectable histological changes is the origin of controversy regarding its normal role in physiology and pathophysiology. APOPTOSIS EN NEFROLOGIA Activation of a genetic program for cell death The development of Caenorhabditis elegans, a nematode, requires the formation of 1090 cells while 131 undergo programmed cell death. The study of the genes involved in apoptosis in this model has originated concepts that have been later applied to amniotes. Two lethal genes have been identified in C. elegans, ced-3 and ced-4, which need to be expressed for cell death to occur, and a survival gene, ced9, which has to be suppressed 6,20. Recent studies have also identified genes that are i n v o l v e d in apoptosis in amniotes (table ll). Molecular studies in tumors have uncovered genes, such as bcl-2 and p 5 3 21,22, whose alterations result in excessive or defective function that promote tumor growth by decreasing the Iikelihood of apoptosis. O t h e r s have been discovered by chance, like Fas/APO-1, first identified with a cytotoxic monoclonal antibody 23, 24. Finally, a systematic approach, comparing the content of cDNA libraries or antigen expression between apoptotic and nonapoptotic cells has yielded a number of genes or proteins, such as tcl-30, the function of which remains unclear 25. We will limit our discussion to genes which have been shown to be expressed by renal cells, drawing on some data from our own experience (figure 2). 12345678 872 b p 603 bp 310 b p Figure 2.-RT-PCR of apoptosis related genes in renal cells. cDNA was made by reverse transcription of murine tubular epithelial cell mRNA. PCR was performed employing primers derived from the published gene sequences. The PCR products (shown in the figure) were cloned into the PCRII vector and sequenced to confirm their identities. Northern analysis of RNA from renal cells and kidney revealed the presence of adequate sized transcripts. Lane 1, bcl-2; 2, bel-x; 3, bax; 4, p53; 5, molecular weight markers; 6, Fas; 7, tcl30; 8, clusterin. Table ll. Apoptosis-related genes l l l Prevent apoptosis: bcl-2, bcl-xL. Promote apoptosis: - Receptors: Fas, 55pTNFr, 75pNGFr. - Transcription factors: p53, c-myc, c-fos, c- jun.. - Intracellular proteins: bax, bcl-x.$ IL- 1ß-convertase. Role in apoptosis not well characterized: tcl3O, clusterin, c-fes, WT-1. Most of the genes under discussion have been well studied in neurons, Iymphohemopoietic, and tumor cells. It remains to be demonstrated that their role in apoptosis is similar in renal cells. There is, however, evidence that the basic mechanisms of apoptosis are well preserved among species, and probably among different organs of the same individual. For instance, the human bcl2 gene protects C. elegans cells from apoptosis 26. Protective genes: bcl-2 and bcl-X The activation of bcl-2 and the bcl-xL isoform of the bcl-x gene prevents or retards apoptosis. bcl-2. bcl-2 encodes a membrane-bound protein expressed on mitochondria and other intracellular mem branes 27, 28. It belongs to a recently identified fámily of interactive apoptosis-related proteins which includes bcl-x, bax and MCL 1 29. bcl-2 appears to function by blocking lipid peroxidation and/or inhibiting the production of reactive oxygen species 30, 31. The bcl-2 gene was identified in the locus of a t(14;18) (q32;21) translocation present in 85% of human follicular lymphomas 21. This translocation results in the uncontrolled expression of a chimeric protein which retains the survival function of bcl-2 21. It is thought that this protein may confer a vital advantage to oncogenic clones, and in conjunction to other mutations, could lead to Iymphoma. Excessive expression of bcl-2 has also been described in epithelial neoplasia 32. bcl-2 overexpression confers protection against apoptosis, and interference with bcl-2 function predisposes to apoptosis in several in vitro and in vivo systems 6, 8-10, 33-39 Specifically, bcl-2 protects from c-mycinduced apoptosis 33, glucocorticoid-induced apoptosis in Iymphocytes 9, Fas/TNFa induced cell death 34,35 and some, but not all, models of growth factor deprivationinduced apoptosis 8 10, 36-39. bcl-2 also protects against death induced by oxidants 30 and physical factors such as heat shock and irradiation 6. However, it offers no protection agdinst cytotoxic T cell killing 40. The over393 A. ORTIZ-ARDUAN Y E. G. NEILSON. expression of bcl-2 in B-lymphocytes of transgenic mice results i n the abnormal survival of autoreactive clones as well as the development of a proliferative glomerulonephritis 41. Interestingly, bcl-2 mRNA levels in circulating Iymphocytes have been reported to be elevated in patients with systemic lupus erythematous 42, although this may just represent the state of activation of those lymphocytes 43. bcl-2 may also play a physiological role, as suggested by its elevated expression during development 44,45 as well as in adult, long-lived c e l l s such as neurons and memory Iymphocytes 46. bcl-2 has a further potential role in renal development, as mice carrying a targeted mutation in bcl-2 develop polycystic renal disease that progresses to renal failure 47. We have recently demonstrated that different murine renal cells, including mesangial cells, tubular epithelium, fibroblasts and metanephric stem cells, exp r e s s bcl-2 mRNA and the protein product is detectable by indirect immunofluoresence 48. bcl-2 mRNA is also present in blood-free adult mouse kidney. As has been previously described in other organs, renal cells and kidneys express bcl-2 mRNA transcripts of severa1 sizes (7.5, 4.1 and 2.4 kb), the 7.5 kb transcript being the most abundant. bcl-x. Bcl-x has recently been described as a new member of the Bcl-2 family of proteins 43. In humans different alternatively spliced isoforms of bcl-x protect from (bcl-xL) or predispose to (bcl-xS) apoptotic cell death in growth factor-deprived cells. The apoptotic promoting action of bcl-xS seems to be related to the inhibition of bcl-2 action. We have demonstrated that murine kidneys and renal cells incultures express two bcl-x mRNA transcripts, detectable by northern hybridization, the smaller one having a shorter half life. Both murine transcripts hybridize to a human bcl-xL specific probe, although their function remains undefined. Genes that promote cell death: p53, c-myc, c-fos/ c - j u n , b a x and IL- 1 b-converting enzime The- p53 gene is required for certain modes of apoptosis. Excessive c-myc expression results in dependence of external factors to prevent cell death. cfos and c-jun have a role in growth factor deprivat i o n - i n d u c e d apoptosis. bax is an endogenous antagonist of bcl-2 IL-1 P-converting enzyme is the mammalian homologue of the death gene ced-3. p53. p53 is a nuclear phosphoprotein with characteristics of a transcription factor, whose action is mediated by binding to DNA and other proteins 22,49. p53 is the most frequently mutated or deleted gene in solid neoplasia, and is inactivated by several oncoviral pro394 teins 22. There has been a misunderstanding of the physiological role of p53 because the original functional studies were performed with mutated p53. It is now clear that wild-type p53 is able to induce apoptosis 7. p53 is required for apoptosis induced by radiation and DNA damaging drugs, but not for dexamethasone-induced apoptosis in thymocytes 50, 51. Moreover, apoptosis induced by p53 in leukemic cells, is partially blocked by growth factors such as IL-6 52. The adenovirus El/B protein, which blocks p53 function and inhibits p53-mediated apoptosis 53, also protected against Fas and TNFa-mediated cell killing 54, suggesting a role for p53 in receptor-initiated apoptosis. p53 is highly expressed in the developing kidney, and is also present in most adult tissues 55. Wowever, the possible role of p53 in renal cell death is unclear. Studies in other organs súggest that p53 may play a role in non-neoplastic processes. p53 immunoreactivity is increased in soft tissue and cervix inflammatory lesions 56,57, as well as in relation to neuronal cell death during experimental cerebral ischemia 58. c-myc. c-myc is a proto-oncogene which functions as a transcription factor 59. c-myc activity depends on the formation of a heterodimer by binding to another intracellular protein, Max. Max is always found in excess relative to c-myc, and c-myc levels determine the formation and activity of the heterodimer 59. cmyc had been traditionally considered a stimulatory factor for cellular proliferation. In fact, c-myc transgenic mice develop cystic kidney disease and hyperplasia in glomerular and tubular epithelium 60. Recent studies have modified this concept. In fibroblasts the uncontrolled expression of c-myc increases cellular proliferation in the presence of growth factors, but in cells deprived of them, it induces apoptosis 61. Growth factor deprivation in IL-3-dependent cells is associated with decreased c-myc expression, anb if c-myc is overexpressed, growth factor deprivation-induced apoptosis is accelerated 62. Moreover, c-myc antisense oligonucleotides prevent anti-CD3-mediated activation-induced apoptosis in T cell hybridomas 63. It is quite possible that c-myc activates a common genetic program that may determine both cell division and cell death, The presente or absence of additional signals may determine the cell fate. This may be related to the activity of bcl-2, which inhibits c-myc-mediated apoptosis 33. The relationship between c-myc-induced cell death and its capacity to activate the transcription of p53 is not clear 64. c-myc may also protect from cell death under certain conditions. During steroid-induced killing of T cells, c-myc transcripts are down-regulated, and transient expression of c-myc protects cells from death 65. c-fos and c-jun. fos and jun homo or heterodimerize to form the AP-I transcription factor. The expres- A. ORTIZ-ARDUAN Y E. G. NEILSON. Ipr/lpr mice (manuscript in preparation). The fact that endotoxin increases the expression of a receptor capable of inducing cell death may provide new insights into the pathogenesis of organ dysfunction in septic shock. Specifically, the mechanisms of renal f a i l u r e in sepsis are poorly understood. Although TNFa has been considered a primary mediator of endotoxin-induced organ damage, mice lacking the 55 Mr TNF receptor are still susceptible to endotoxin, but not to TNFa/lLl P-mediated death 96. In this regard, Fas, a receptor similar to the 55 Mr TNF receptor, is an excellent candidate to explain why mice carrying targeted mutations in the TNF receptor are not completely protected from the deleterious effects of LPS. Fas has also been shown to be expressed in metanephric stem cells under the regulation of TNFa (unpublished data). These data seem to fit with the recently proposed general role of macrophages, possibly through engagement of Fas or similar receptors, in tissue remodeling during organogenesis 97. ce, as well as in the glomeruli of LPS-treated mice 111. Thus, clusterin deposited in the glomeruli may be of local origin. We also demonstrated that mesangial cells in culture express clusterin mRNA 111. However, in these cells, clusterin expression was decreased by conditions that induced cell death compatible with apoptosis, such as TNFa and serum deprivation, as well as by ylFN, which did not kill the cells. This indicates that clusterin expression may be regulated in glomeruli by cytokines, independently of apoptosis. Other authors have raised questions about the relationship of clusterin and apoptosis in other experimental systems and have proposed a possible role for c l u s t e r i n in cell differentiation 112. These authors found abundant clusterin mRNA in newly polarized cells of comma and S-shaped bodies, but not in uninduced or non-polarized metane phais mesenchyma. In keeping with these findings, our recently characterized metanephric stem cell line MMR 113 did not express clusterin mRNA, unlike renal fibroblasts, mesangial and tubular cells 111. Other apoptosis related genes t c l 3 0 . tcl30 encodes a 2.4 kb transcript discovered during a comparison of cDNA Iibraries from normal and apoptotic thymocytes 98. Apoptotic thymocytes express high Ievels of tcl30, which in vivo was detected only in the thymus. The possible role of tcl30 in apoptosis is unknown. However, it is interesting that it encodes a membrane protein with a vitronectin-like domain, and that the vitronectin receptor has been implicated in the phagocytosis of apoptotic debris by mesangial cells and macrophages 99. We have recently shown by RT-PCR and northern blot that tubular epithelial cells express tcl30 mRNA. Clusterin (complemen t Iysis inhibitory pro tein). Clusterin is a multifunctional glycoprotein, which has received a variety of names (table III) 100. Clusterin is associated with apóptosis, although its role in this process, if any, is unknown. Together with c m y c , clusterin is one of the few apoptosis-related genes which. has been extensively studied in the kidney. Clusterin expression is increased in several models of acute renal failure, and in chronic models, such as subtotal nephrectomy and polycystic diseases 101-106, some of which have been associated with the occurrence of apoptosis. Clusterin is also found deposited in the glomeruli in immune glomerulonephritis 107-109, and one of its best characterized functions is the inhibition of the activity of the terminal complement complex (C5 b-C9) 110.. We recently observed increased levels of clusterin mRNAin the kidneys of mice with anti-CBM disease and in MRL-lpr/lpr lupus mi396 Table III. Clusterin Synonyms: l SGP2; sulfated glycoprotein 2 (rat). l TRPM: testosterone repressed protein messenger (rat). l GP80: glycopoprotein 80 (canine MDCK cells). l SP40,40: serum protein 40,40 (human membranous nephrol l APO j (human). Glycoprotein III (bovine). path y) l Functions: l Binding and inactivation of C5b-C7 complement complexes. l Increased in relation to apoptosis in vitro and in vivo. l Apolipoprotein. l Neuroendocrine cell granule component. Other genes. Reports on the gene regulation of apo lptosis-are frequent. Recent papers have reported that antisense oligonucleotides complementary to the c-fes oncogene induces apoptosis in HL6O cells 114. Activated T24-ras and v-ab/ have also been shown to protect from apoptosis, at least under certain conditions 29. Apoptosis in pathophysiology The apoptosis is a normal biologic process in most embryonic and adult tissues. Together with mitosis it regulates the number of cells in each tissue at each stage of development (figure 3). In fact, cell proliferation and cell death seem intimately related with e x a m p l e s suggestive of a common activation pathway that may result in one or the other 61,63, depen- APOPTOSIS EN NEFROLOGIA sion of c-fos and c-jun is rapidly and transiently induced upon growth factor deprivation in IL-2 and IL-6dependent cell lines, preceding apoptosis 66. Moreover, antisense inhibition of either of them protected against death 66. An association between c-fos expression and apoptosis in transitory embryonic structures has also been noted 67. fos and jun may also be involved in apoptosis regulation through the formation of the transcription complex NF-AT. This complex requires a nuclear formation, which includes c-fos and c-jun 68, and a component that has to be translocated from the cytoplasm. Cyclosporin A, which blocks activation-induced apoptosis in T cells 69, also blocks the translocation of the cytoplasmic component of NF-AT, probably by inhibition of calcineurin activity 70. bax. Bax was described in August 1993 as a member of the Bcl-2-like family of proteins 71. Bax forms heterodimers with Bel-2. In cells transfected with both bcl-2 and bax, the bcl-2/bax ratio appears more important in determining the susceptibility of the cell to growth factor deprivation-induced apoptosis than either of the two alone: excessive bax predisposes cells to apoptosis under these conditions, but not in the presence of growth factors. The original report of this finding suggested that the murine kidney expresses two bax transcripts of 1.5 and 1 kb abundantly. In o u r own laboratory we have found the 1 kb bax transcript to be preferentially expressed by mesangial cells, proximal tubular cells and metanephric stem cells as well as by whole kidney. IL-1b-converting enzyme. CED-3 is required for cell death to occur in C. elegans. Mammalian IL-l /3converting enzyme is highly homologous to CED-3 72. In fact, overexpression of IL-l P-converting enzyme has been shown to induce cell death in rat fibroblasts 73. This effect was prevented by mutating the gene, by the IL-l p-convertase antagonist crmA gene, and by bel-2. It had already been suggested that this enzyme might have other roles besides cleaving IL1 B, as it is expressed by cells that lack the cytokine, and it i s activated in macrophages undergoing apoptosis, but not in those undergoing necrosis 74. Other ced-3 related genes such as Nedd-2, may also have a role in apoptosis. Receptors that induce apoptosis: Fas and the TNF receptors family of proteins Fas (CD95) is a cell membrane protein, and a member of a family of receptors that includes both TNF receptors 75, 76, the low affinity NGF receptor 77, CD40 78, OX40 79, CD27 80, and CD30 81. This protein family is defined by similarities in their extracellular domains that are absent in the intracellular portion of the molecule. Several of these proteins modulate the occurrence of apoptosis. Both the activation of Fas and the 55 Mr TNF receptor induce cell death with features of apoptosis through a relatively conserved intracellular domain 82,83. However the 75 Mr NGF ? receptor lacks that domain, but still induces apoptosis when not bound to NGF: unlike Fas and the TNF receptor, it is NGF's nonoccupancy of its receptor which induces apoptosis 84. Fas was initially described by two independent groups which developed monoclonal antibodies (APOI and Fas) that were cytotoxic to human cells 23,24. A recently described anti-murine monoclonal antibody also induces apoptosis in vitro and in vivo 85. Fas is another example, along with bcl-2, of apoptosis-related genes whose alteration leads to an immune-mediated glomerulonephritis 41,86. The genetic defect of Fas in MRL-lpr/lpr lupus mice results in the survival of autoreactive T cell clones and the appearance of a generalized autoimmune disease. This defect consists in the insertion of an endotransposon in the second intron of the Fas gene, leading to abnormal splicing of the gene, with nearly total absence of normal sized transcripts and the expression of low levels of abnormally sized messages in the thymus and liver 87-90. The abnormal transcripts contain a 168 bp endotransposon fragment within the coding region of the gene. Interestingly, the normal splicing pattern can apparently be restored in the thymus of TCRP transgenic MRL-lpr/lpr mice 89. This lead us to hypothesize that certain, as yet unknown stimuli, may induce the expression of normal transcript in the kidney of these animals. The endogenous ligand for Fas has recently been characterized and shown to induce apoptosis 91. The Fas ligand is thought to be encoded by the gene defective in gld lupus mice 92. Studies based on this assumption suggest a role for Fas in T cell-mediated cytotoxicity 93. Cytotoxic T cells can play a pathogenic role in experimental and human renal diseases 94, and Fas may participate in this process. We have recently shown that murine mesangial cells and tubular cells express Fas mRNA under the regulation of cytokines and endotoxin 95. Cytokines thought to play a pathogenic role in kidney damage, such as TNFa, IL1 p and ylFN, increased levels of Fas mRNA and expression of the Fas receptor in renal cells, and levels of Fas mRNA, for example, are increased in anti-GBM disease 95. Endotoxin is also a potent inducer of new Fas transcripts by renal cells in culture and by the kidney, liver and lung in vivo 95. Moreover, endotoxi n resu Ited in expression of near normal levels of normal sized Fas message in the kidneys and liver from MRL395 APOPTOSIS EN NEFROLOGIA The number of cells in mature tissue depends on the balance between mitosis and apoptosis yq=CELL DEPLETION HOMEOSTASIS CELL ACCUMULATION Figure 3.-The homeostasis of cell number depends on the balance b e t w e e n cell division (mitosis) and cell death (apoptosis). An excess of cell mitosis, especially if associated to a decreased rate of cell death, may result in disease characterized by excessive cell a c c u m u l a t i o n , such as proliferative glomerulonephritis. An increased rate of cell death, especially if not compensated by i n c r e a s e d ce ll division, may result in cell depletion, such as chronic renal atrophy. presence of adequate inducers: mice carrying targeted mutations of WT-1 suffer from renal agenesis as a result of massive apoptosis of the metanephric blastema 117. Growth factors such as NGF and EGF can rescue metanephric cells from apoptosis 19, 116, and antisense oligonucleotide experiments sugges t that the 75 Mr NGF receptor participates in kidney development 118 Embryonic human kidneys also express high . levels of the T cell and monocyte chemoattractant R A N T E S (A. Krensky, personal communication). Although its role during development is as yet unclear, this fits with the recently proposed essential role of macrophages during embryonic apoptosis. Moreover, metanephric stem cells can be induced in culture by macrophage products, such as TNFa, to express transcripts that encode for an apoptosis transducing receptor, Fas (unpublished observation). Table IV. Apoptosis in physiopathology Partial list of physiological processes characterized by the occurrence o f a p o p t o s i s . l Embryonic development. Tissue turnover: - Keratinocytes. - Intestinal epithelium. Degeneration of hormone-dependent tissues after removal of trophic hormones: - Thyroid, adrenals. - Prostate, breast, endometrium. Regulation of the immune response: - Negative selection. - Control of peripheral immune response (activation induced apoptosis). - CD4 and CD8 J cell cytotoxicity. Hematopoiesis: - Regulation of circulating blood cell number. - Generation de erythrocytes by normoblast apoptosis. ding on the instructions of endogenous or exogenous factors. Inhibition of apoptosis, for example, may be viewed as cell proliferation in in vitro studies. Table IV summarizes some of the physiological processes in which a role for apoptosis has been suggested. The rapid tissue turnover and modification that characterizes embryonic development would not be possible without apoptosis. The role of apoptosis in the elimination of unwanted cells during development may be more fundamental than the previously thought. A recent study has suggested that prevention of apoptosis, per se, in the absence of differentiating factors, may be enough for differentiation to proceed 115. In some parts of the central nervous system, for example, up to 60% of the neurons that are formed die by apoptosis 17. There is limited information on apoptosis and kidney development. Developing kidneys are known to contain apoptotic cells 19,116 and to express high levels of several of the apoptosis-related genes, like bcl-2 and p53 44, 45, 55. In fact, mice carrying targeted mutations of bcl-2 have neonatal polycystic kidney disease with persistence of immature cells 47. Our own data indicate that bax and bcl-x are expressed by embryonic kidney and metanephric stem cells in culture. The relative levels of the two murine bcl-x transcripts in the kidney change during development with lower amounts of the larger transcript found in the adult kidney (unpublished observation). Uninduced mesenchyme dies by apoptosis 116, and WT-1 is required for the prevention of apoptosis even in the l l l l Alterations in the frequency of apoptosis can cause disease characterized by insufficient or excessive number of cells. The loss of function of genes that induce apoptosis or the increased expression of genes that prevent apoptosis can result in autoimmunity, excess of autoreactive cells (bcl-2, Fas), or neoplasia ( bcl-2, p53) 21,22, 41,86. An excess of apoptosis has been implicated in the neuronal loss of Alzheimer's and Parkinson's diseases, and in the disappearance of T lymphocytes following HIV infection 6 119, 120. Altered regulation of apoptosis may also result in persistent inflammation or fibrosis, if the numbers of leukocytes and fibroblasts in a site of injury are not adequately control led 121 . A. ORTIZ-ARDUAN Y E. G. NEILSON. Apoptotic and renal disease Studies on apoptotic and renal disease are recent and more abundant in abstracts presented at meetings than in the published Iiterature. Published studies have demonstrated the existence of apoptotic morphology and the typical pattern of DNA degradation in several renal diseases, although there are numerous unanswered questions regarding the factors that induce and prevent apoptosis, its gene regulation, its relative contribution to renal damage, and the possible therapeutic value of the modulation of apoptosis. Table V summarizes renal diseases in which apoptosis may play a role. Table V. Possible participation of apoptotic in renal d iseases Metanephric development. Acute renal failure: toxic, ischemic, obstructive. Chronic renal failure: renal atrophy and interstitial fibrosis. Inflammation: loss of parenchymal cells, resolution of inflammation and of cell proliferation (hyperplasia). Autoimmunity and transplantation: regulation of the inmune response. Polycystic renal disease. Nephrotic syndrome. Diabetic nephropathy. Tumors. There are four general aspects to the relationship of apoptosis to rena l disease: Apoptotic as renal cells a mechanism o f depletion o f intrinsic Apoptotic induced by a reduced supply of growth factors, or the presence of soluble or membrane bound products of macrophages or T cells may play a role in the loss of parenchymal cells observed in ischemic, toxic and inflammatory processes. Apoptotic and renal failure. There is evidence for the involvement of apoptosis in both the acute and c h r o n i c loss of renal parenchyma 14, 122,123. Internucleosomal DNA degradation and the presence of apoptotic bodies have been reported to occur in both ischemic acute renal failure and chronic ischemic atrophy 14, 123. In fact, after brief ( <45 min ischemia and reperfusion apoptotic figures were detected in rat kidneys at 12 hours the classic internucleosomal dna fragmentation was also evident 123 chronic could be observed up to 28 days 14 these conditions there is ongoing necrosis relative contribution of two mechanism death cell loss may depend on 398 balance regulatory events for obscure reasons apoptosis seems specially frequent postransplant acute renal failure 124 evidence has been found toxic obstructive r e n a l 48 122 125 cortical atrophy post-papillary 126 disease caused by subtotal nephrectomy 127 we others have demonstrated that sensitive southern blot technique permits identification degradation even normal kidney 128 however short vivo half life makes it challenge study its participation p o t i c indices below 1 5 result 50 cells during central nervous system development 17 moreover injury detachment an early event can shed into tubular lumen while meaning presence viable urine not yet fully understood 129 hypothesized destined undergo are rescued from vitro culture growth factors 13 little known about precise damage direct consequence exposure noxious stimuli occur proliferating compensatory fashion after absolute or deficits this setting might contribute persistence delayed recovery physiologic check reverse exaggerated proliferative response shimizu et al 60 minutes peak first 48-72 h second bigger 7-14 when necrotic tubules had completely relined hyperplastic epithelium elevated indexes persisted 4 months proposed participate regression hyperplasia induced lead gentamicin 130 131 issue addressed experimental but relevant clinical practice effect repeated low level insults such as recurrent episodes hypoperfusion hypotension infection ability withstand theory select rather than because mildness insult repetitive nature which would deprive replacement parenchymal inflamma tion role en nefrologia inflammatory reviewed several cytokines mediators released macrophages induce other types both cd4 cd8 cytotoxic lymphocytes target 132 133 donor untreated rejection liver transplantation shows histologic features 134 possible nephrotic syndrome generally accepted idiopathic podocyte information regarding denudation basement membrane reported severa1 forms human including minimal change focal segmental glomerulosclerosis glomerulonephritis diabetic nephropathy 135 morphological originally interpreted those coexistent missed due already commented difficulties identifying further studies applying newly developed techniques situ detection 136 should performed cytotoxicity adriamycin puromycin aminonucleoside cultured glomerular epithelial 137 related their apopto138 139 although unclear why podocytes sis especially effects interesting depends activation p53 only express wt-i adult 140 interaction between increases transcriptional activity 141 te more rapidly survive longer obtained healthy 142 similar observations made with mesangial animals recently resolution proliferation characteristic anti-thy-l nephritis excessive numbers 143 occurrence 144 abnormaiities apoptosis-related genes polycystic transgenic mice overexpressing c-myc suffer over-expression cpk 145 interestingly bcl-2 protects c-myc-induced 33 still lack functional bel-2 results neonatal 47 displayed cysts tubulointerstitial lesions immature mitotic suggesting increase rate leads alternatively premature compromise differentiation 115 theoretically over expression bax endogenous antagonist abnormal levels mrna unpublished observation decreased neoplasias altered some carcinomas regulation recent data suggest leukocytes site eliminated if adequate microenvironment cell-preservation 146-148 clearance mechanisms regulating interstitial inflammation 121 process support hypothesis capacity engulf neutrophils 149 well visualization debris inside 150 survival signals help understand certain differ intrinsic door open manipulation through modulation leukocyte without damaging 399 reduced f accumulations somatic nephropathies defined accumulation examples hypercellularity increased number fibroblasts fibrosis advanced diseases cystic explain alterations nephropath s possibility directly studied indications so fibrotic accumula- ortiz-arduan y g neilson parenchyma fact dual action apoptosis: tnfa example prevents monocytes killing 148 point comment harbored notion itself does generate conceivable massive occurs organ physiologically prepared dying phagocytised adjacent latter dysfunctional they exposed same pernicious toxins disintegration bodies release non-specific proinflammatory lysis w takes place upon injection anti-fas antibody 85 genetic programs activated provide specific chemotactic substances phagocyte recruitment new surface determinants recognition phagocytosis scattered undergoing il-1 74 transcription luciferase construct under control monocyte chemokine scya rantes promoter doses 151 152 widespread unwanted associated leveis krensky personal communication chemoattractant je peaks 24-48 remains long 168 153 invading characterized absence tissue previous clear connection betweén infiltrates indispensable eliciting 97 exact primary engagement receptors fas complement derived means macrophage poorly ne thymus periphery 2 details involvement beyond scope review noted autoimmune drugs used therapy immune-mediated transplant involve lymphocyte 69 154 155 regulate general withdrawal inducing idea viability externa1 increasingly 20 factor requirement vary type status additional external neuron hemopoietic much less needs shown serum deprivation 156 account promoting unknown egf igf-i 157-159 administration improves evolution 160 161 decreases protect against systems 162 163 metanephric proximal 19 pdgf main element responsible mitogenic 164 165 temporally appearance following 158 part developing oligodendrocytes 18 better understanding tissues therapeutic interest recombinant available least addition 24 rescue otherwise die modulator immune growing b d ev idence plays fundamental immu400